Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf

1 Publication
Quantification of the metabolic uptake rate in whole body F-18 FDG PET: a comparison between dynamic and static imaging
van den Hoff, J.; Hofheinz, F.; Weise, R.; Maus, J.; Preuß, R.; Burchert, W.;
Dynamic whole body (DWB) FDG PET has become available with the recent introduction of the "flow motion" package by Siemens which offers fully automated generation of parametric images of the metabolic uptake rate K[m]. While this approach is superior to SUV-based quantification, it requires substantially more scan time and reduces patient throughput. On the other hand, the tumor to blood standard uptake ratio (SUR) has outperformed SUV in several clinical studies, which is attributed to a very high correlation between SUR and K[m]. However, direct evidence for this correlation is scarce. Our study compares K[sur], the SUR-derived "static" estimate of K[m], with the Patlak-derived K[m].

Altogether, 12 oncological patients are enclosed and scheduled for DWB PET. Parametric images of K[m] and K[sur] are computed using the vendor provided Patlak procedure and in-house software, respectively. Evaluation of the voxel intensity correlation between both parametric images as well as ROI-based analysis is performed.

So far, 4 patients have been evaluated. K[m] and K[sur] are highly correlated (R^2=0.97) in areas with nonnegligible irreversible uptake (so far we investigated the range K[m]=[0.01,0.07]). K[m] and K[sur] images have very similar image contrast between such areas. Minor contrast differences exist in healthy soft tissue and regions where the Patlak approach is invalid (notably the liver). The relative scale factor between both parametric images is 0.86.

Our results suggest that SUR and K[sur] are proportional surrogates of true K[m]. The unquestionable potential of DWB PET might thus be more relevant for applications beyond oncological FDG PET. To test this conjecture, comprehensive studies in homogeneous patient groups are required in order to compare the prognostic value of K[m] and SUR/K[sur] in the context of survival analysis
  • Lecture (Conference)
    Jahrestagung der DGN, 03.-06.04.2019, Bremen, Deutschland

Publ.-Id: 29096 - Permalink